Gene Expression Distinguisher Identification in Mixed Biological Samples

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Biological samples often contain multiple cell types, leading to substantial background noise and overlapping data, making it challenging to accurately analyze tissue composition and expression levels, which is crucial for diagnostic purposes and understanding disease mechanisms.

Innovation Solution

A system and method that generates a joint expression matrix from biological sample data, normalizes rows, and selects gene expression distinguishers based on magnitude and orthogonal projections, allowing for the identification of key genes without prior knowledge of biological processes, thereby distinguishing cell types and biological processes in mixed samples.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional expression measurement analysis is performed on heterogeneous tissue samples, then diagnostic information can be obtained, but substantial background noise and overlapping data from multiple cell types mask accurate analysis of tissue composition and expression levels

Engineering Contradiction:
Improveaccuracy of expression level measurementVSAvoidbackground noise from mixed cell types
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent segments the heterogeneous expression data into distinct cell type-specific expression profiles by identifying expression distinguishers that characterize different cell populations. The convex analysis method decomposes the mixed sample expression matrix into pure cell type expression signatures and their relative abundances, effectively separating the contribution of each cell type to the overall heterogeneous signal.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces mathematical modeling and convex analysis as intermediary tools to bridge the gap between heterogeneous bulk tissue expression data and cell type-specific expression profiles. These computational methods serve as mediators that extract meaningful cell type-specific information from the mixed signal without requiring physical separation of cell types.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If computational methods are used to identify cell type-specific expression profiles from mixed samples, then accurate tissue composition estimation is achieved, but the complexity of data processing and analysis increases

Engineering Contradiction:
Improveaccuracy of tissue composition estimationVSAvoidcomputational complexity of analysis method
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary identification of expression distinguishers and establishment of convex analysis frameworks before applying them to specific heterogeneous samples. By pre-defining the mathematical structure and identification algorithms, the method reduces the computational burden during actual analysis while maintaining high precision in tissue composition estimation.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3590059B1Method for identifying expression distinguishers in biological samples
Publication Date: 2023.09.13 GLOBAL LIFE SCI SOLUTIONS OPERATIONS UK LTD
  • EP3590059B1 patent drawingFigure 1~2
  • EP3590059B1 patent drawingFigure 3
  • EP3590059B1 patent drawingFigure 4~5

AI summary

The present techniques provide techniques for determining gene expression distinguishers of biological samples using expression data that comprises signal intensity of signal generators with binding specificity to target molecules. Multiple samples may be analyzed to determine gene expression distinguishers that may be used for identifying cell types or understanding the mechanism of disease progression.